Type 1 diabetes mellitus (T1DM) results from the immune- (specifically T cell-) mediated destruction of the body's only cells capable of physiologically regulated insulin secretion, the pancreatic beta cells. Beta cells comprise an estimated 2% of the pancreatic cell number and are grouped together into cell clusters (along with blood vessels and other cells making hormones like glucagon, somatostatin, pancreatic polypeptide) into "mini-organs" called the islets of Langerhans. The islet as "mini-organ" concept is apt for at least 2 reasons: (1) while the cellular organization is different for different species, the various islet cell types are nevertheless organized in a very orchestrated way, and (2) islets consume more pancreatic blood flow (about 20%) than their small mass would suggest. Thus islets are important if incompletely understood structures with unique anatomical, physiological, and immunological features; and they represent an Achilles' heel for individuals destined to develop T1DM. This project has multiple components, to: (1) characterize isolated islet structure, function, and quality, (2) test, using a non-human primate islet transplant model, important pre-clinical questions, (3) perform human clinical islet transplants using carefully planned protocols, (4) develop clinical assays for characterizing islet function pre- and post-transplant, (5) develop a renewable islet source, and (6) test novel ways of preventing islet allograft rejection following transplant. Beginning in 7/99, in collaboration with the Clinical Center's Department of Transfusion Medicine/Cell Processing Unit and Drs. Ricordi and Kenyon (University of Miami's Diabetes Research Institute), islets have been isolated from both human and non-human primate pancreata. Many of these islets have been shared with various intra- and extra-mural collaborators pursuing shared goals as enumerated above. To enable these isolations, we established 2 separate laboratories, one for human glands and one for animals, and we trained a team of technicians. We perform the standard in vitro islet function assays (islet insulin release in low- and high-glucose media, viability assays, and this year we've tested an oxygen consumption rate assay). In addition, we've initiated efforts to test all islet preps using the difficult but important gold-standard islet function assay (i.e. transplanting the islets into diabetic NOD-scid mice). We also performed new in vitro islet function assays and made, or further pursued, several novel observations including: a) using RNA expression microarray analysis we found that islets express, and in a highly regulated, glucose-responsive way, a TGF-beta gene that appears to play a role in beta cell survival, b) identification of an insulin RNA splice form translated into protein more efficiently than the native splice form, and we found that this splice form is highly expressed in islet tumors, and c) data demonstrating that islets produce the hormone resistin, previously thought to be expressed only in fat. Several collaborators are attempting various techniques to culture islets in vitro so as to increase islet function and/or number, and Branch investigators have played a leadership role in the international group called the Beta Cell Biology Consortium. In addition, we have continued a detailed follow up of the non-human primates given an islet transplant and found a previously unknown mechanism whereby the transplant improves glycemia control; i.e. by inhibiting the over-expression of the hormone glucagon displayed by patients with diabetes (manuscript submitted). Last, as clinical islet transplant experience has revealed a high incidence of either bleeding or clotting complications, Branch investigators have developed an animal model to further explore the sequella of injecting isolated islets into the blood stream. We are studying this model so as to test ways to mitiigate the clotting/bleedding complications. All these studies are designed to support NIH islet transplant clinical trials. Of course, the most immediately relevant product of our research is the knowledge gained from our clinical islet transplant experience which included 6 patients with long-standing and difficult to control T1DM. Detailed metabolic testing of the patients found they had basically normal insulin sensitivity, but even the insulin independent patients had only marginal islet function and imperfect glycemia control. Due to our belief that the factors limiting islet transplantationn (primarily the inadequate donor islet supply, and imperfect immunosuppressive regimens) were not being effectively and most safely addressed by the solitary islet transplant protocol, we suspended patient accrual, but have published several observations from our experience. For instance, we this year published an analysis strongly suggesting that pancreas transplantation and the subsequent post-transplant anti-rejection therapy, when performed for individuals with diabetes and preserved kidney function, resulted in worse survival than that observed in patients receiving more traditional diabetes therapies- and for up to 4 years post-transplant. This disturbing observation has important consequences beyond the islet transplantation research field as pancreas transplantation is performed widely in the U.S. and our manuscript suggests a reappraisal of the therapy's application. Indeed, in addition to our published scientific manuscripts, two widely respected scientific journals have published commentaries based upon our work (Science, Vol 306: pp 34-37, 2004; and JAMA, Vol 290: pp 2861-2863, 2003). On a more positive note, we have observed that a surprising number with even long-standing and difficult to control T1DM maintain a limited capacity for endogenous insulin production. This observation, which is inconsistent with the most widely held model for T1DM, has prompted several new protocols designed to test whether pancreatic islet function might be promoted in the patient with T1DM. For instance, in FY2005 we will develop, using our non-human primate model, a technique for differentiating islet function emanating from the native as opposed to a transplanted pancreas, then apply that technique to study patients with long-functioning pancreas allografts. In addition, as described in Project Number DK062002-05, we are testing whether combination therapy with insulin (to maintain near normal glycemia control), an experimental agent called Exendin-4, and/or immunosuppression can promote pancreatic islet functional recovery in patients with long-standing T1DM. Last, plans are underway to develop techniques to follow islet mass in vivo, as described in Project Number DK062004-05, and any promising results from those studies will be transitioned into the clinic under this and other Branch projects.